Sunteți pe pagina 1din 13



David Crystal

Many of the terms used in the study oflanguage are 'loaded', in that
they have a number of different, sometimes overlapping, sometimes
contradictory and controversial senses, both at popular and
scholarly levels. The word STYLE is a particularly good example of
the kind of confusion that can arise. The multiplicity of meanings
which surround this concept - or, perhaps set of concepts - testifies
to its importance in the history of English language studies, and
indicates the magnitude of the problem facing any student of the
subject. On the one hand, there are highly technical definitions of
style such as 'the style of a text is the aggregate of the contextual
probabilities of its linguistic terms' (Enkvist); on the other hand,
there is the loosely metaphorical, aphoristic definition of style as
'the man himself (Buffon). Style has been compared to thought,
soul, expressiveness, emotions, existence, choice, personality, good
manners, fine clothing ... and much more. How, one might well
ask, is it possible to sort out such a semantic tangle? For sorting out
there must be, if there is to be any clear discussion of this unde-
niably fundamental aspect of people's use of language.
One useful way into the tangle is to look at the most important
senses in which the word STYLE is used at the present time, and see
if there is any common denominator, or dominant use. There may
be no single answer to the question, What is style?, but it should at
least be possible to distinguish the main strands of meaning which
would underlie any such answer.
The first, and possibly the most widespread use, is to take STYLE
as referring to the distinctive characteristics of some SINGLE auth-
or's use of language - as when we talk of 'Wordsworth's style', or
make a comment about 'the style of the mature Shakespeare'.
There are a number of different areas of application for this inter-
pretation: for example, we may want to clarifY some comparative
question (as when comparing the 'styles> of two poets in a given

tradition), or we may be concerned with the study of some single In the light of these examples, the term STYLE can be seen to be
author as an end in itself, or again we might be engaged in stylistic applicable, in principle, to a great deal of language use other than
detection work - 'linguistic forensics', as it is somctimes half- literature; and on the basis of this we might well generalize and say
seriously called - as with the investigations into the 'style' of the that style seems to be a concept which is applicable to the language
Pauline epistles, to see whether one man wrote them all. But in each as a whole. The word 'distinctive' has occurred a number of times
of these applications, the primary task is the same: to pick out from already in this chapter. If one of the bases of style is linguistic
the totality of the language that an author has used those features distinctiveness of some kind, then it is very difficult - probably
which would be generally agreed as belonging to him, identifYing him impossible - to think up cases of uses of English in which there is
as an individual against the backcloth of the rest of the language- no distinctiveness whatsoever. Even the most ordinary kinds of
using world. And it is these idiosyncratic linguistic markers which are conversation have the distinctive feature of being 'most ordinary'.
referred to by this first use of the term 'style'. If we beware of the Non-literary uses of language must not be decried simply because
metaphor, 'style is the man' is an appropriate summary ofthe focus of they are non-literary. To refer to such uses as 'style-less' is to beg
this view.
the whole question as to what style consists of, and to ignore a
A second, and closely related use, is to talk about 'style' in a highly important perspective for literary study. Without an
collective sense, referring to GROUPS of literary figures, as when 'ordinary' style, or set of styles, which we are all familiar with and
referring to the 'style' of Augustan poetry, or generalizing about the use, it is doubtful whether we would ever appreciate an ex-
style associated with one particular genre of drama as opposed to traordinary style, as in literary linguistic originality. This is a point I
another. This is a more general sense, obviously, but it is to be noted shall return to later.
that the procedure for arriving at any conclusions in this area is Other senses of the term STYLE may be found, but they take us
precisely the same as in the study of individual authors: distinctive into a quite different dimension. These are mainly variants of a
linguistic features have still to be identified and described - only this sense of style as a 'quality' of expression. When we talk about
time the use of these features is shared by a number of people, and someone or something displaying 'style', we are making an intuitive
are not idiosyncratic in the narrow sense of the preceding paragraph. judgment about a (usually indefinable) overall impression - as when
These two senses are the most common in any discussion about Mr X is said to 'have style', whereas Mr Y has not. This is very near
literature, in view of the emphasis in literary criticism on defining the to the sense of 'style' as 'powers of lucid exposition or self-expres-
individuality of authors and tracing the development of genres; but in sion': Mr Z 'has no sense of style at all', we might say. Then there is
terms of the study of the English language as a whole, it should be a wholly evaluative sense, as when we talk of a style as 'pretty',
stressed that these senses are extremely narrow. They are restricted 'affected', 'endearing', 'lively', and the like. These uses are very
largely to literary English, and to the written form of the language. different from those described in previous paragraphs, as what we
But we can - and do - equally well apply the term STYLE to spoken are doing here is making value judgements of various kinds about a
English, whether literary or not, and to written English which has particular use of language, passing an opinion about the effect a use
nothing to do with literature at all; and it is this more general use of language has had. The difference between the phrases
which provides us with a third sense. For example, when we refer 'Shakespeare's style' and 'affected style', essentially, is that the first
(usually in a pejorative tone of voice) to the 'style' of Civil Service is a descriptive statement, referring to certain features of the Eng-
prose, or to 'business-letter style', or to the 'formal style' in which lish language which could presumably be pointed out and agreed
sermons or proclamations are given - or even to the 'style' of upon in a reasonably objective way; the second is an evaluative
newspaper and television advertisements - we are referring to an statement, where a subjective judgement is passed about some aspect
awareness of certain features of English sounds and spellings, of a use of language, and where we are told more about the state of
grammar and vocabulary, which characterize in a distinctive way mind of the language critic than about the linguistic characteristics
these particular uses. And comparably familiar examples could be of the author being assessed. Any critical task will involve both
cited of people referring to the style of individuals, as well as of elements, descriptive and evaluative, in varying degrees,
groups - 'I do like J olm's lecturing style, don't you?' corresponding to WHAT we respond to and HOW we respond to it.

What must be emphasized is the importance of placing our speaking world is full of people who want to make everyone else
evaluative decisions in a thoroughly descriptive context: value speak as they do, or as Shakespeare did. It is a pity that the fact and
judgments with no 'objective correlative' to support them may give fundamental role of variety in the English language cannot be
us a great deal of personal pleasure, but they do not provide accepted for what it is - an inevitable product of language develop-
anything of permanent critical value. We can only resolve a debate ment.
as to the merits or demerits of someone's style if the parties in the What, then, are these varieties? The kind of variation which people
debate are first and foremost objectively aware of the relevant are most readily aware of usually goes under the heading of
characteristics of the language they are discussing. The descriptive, REGIONAL DIALECT. It is not difficult to cite examples of people who
identifYing task is quite primary, as it provides the basis for the speak or write differently depending on where they are from. This
response which any two critics might be arguing about. Why does X is one of the most well-studied aspects of language variety. The
think that line effective, whereas Y does not? The descriptive major rural dialects of Great Britain have all been studied in some
analysis of a piece of language (1 shall call this, whether written or detail, at least from the phonetic point of view, as have many of the
spoken, a text) is in no sense a replacement for a sensitive response dialects of the United States. Urban dialects - such as those of
to that language, as some critics of a linguistic approach to literature London, Liverpool, Brooklyn, and Sydney - have on the whole
have implied - how could it be? It is simply an invaluable pre- been less intensively studied, but their distinctiveness is as marked
liminary which is likely to promote clear thinking. What such a as that of any rural area. Take, for instance, the language of
descriptive analysis might involve I shall outline below. currency heard in parts of Liverpool a few years ago: og or meg
When such matters are considered, it becomes very clear that (halfpenny), two meg (penny), joey (threepence), tiddler (silver
there is unlikely to be a single, pithy answer to the question 'What is threepenny piece), dodger (eight-sided threepenny piece), sprowser
style'. And perhaps therefore a more constructive question might (sixpenny piece), ocker (shilling piece), and so on. Terms such as
be: 'What is there in language that makes us want to talk about kecks (trousers), jigger (back alley), ozzy (hospital) and sarneys
"style", in any of its senses, at all?' This approach can be revealing: (sandwiches); phrases such as good skin (nice chap), to get a cob on (to
not only does it display the complexity of the concept of style very get into a bad mood) and that's the gear (that's fine); sentences such
clearly; it also integrates this concept with that of 'language' as a as don't youse butt in with the men (don't interfere with what we're
whole, and thus produces a more general characterization than any doing) and I'll put a lip on you (I'll hit you in the mouth): all these
of those so far reviewed. The approach is, briefly, to see 'style' in illustrate clearly the kind of language variation which can only be
the context of the socially-conditioned VARIETIES a language may be explained in terms of geographical place of origin.
shown to possess - and this is the reason for the title of this chapter. Three points should be noted in connection with regional
The idea that the English language can be - indeed, HAS to be - dialects. The first is that this kind of variation is usually associated
seen in terms of varieties is one of the themes underlying the first with variation in the SPOKEN form of the language. The existence of
chapter of this volume. The phrase 'THE English language' is itself a standardized, written form of English, which all people born into
highly misleading, for there is no such animal. If we look at the use an English-speaking community are taught as soon as they begin to
of English in all parts of the world expecting to find identical write, means that modern dialects get written down only by their
sounds, spellings, grammar and vocabulary on all occasions, then introduction into a novel or a poem for a particular characterization
we are in for a rude shock. There is a great deal in common or effect. The speech of the gamekeeper in Lady Chatterley's L(JVer,
between 'American' and 'British' English, for instance - to take one or that of many of the characters of Dickens, or of the 'regional'
example that regularly rears its head in the letter-columns of the novelists such as Joyce, indicates this point abundantly - but even
press - but people are much more aware of the fact that there are here, only the vaguest approximation to the original pronunciation
differences. The English language is not a single, homogeneous, is made. (After all, if we tried to indicate this pronunciation with any
stable entity: it is a complex mixture of varying structures. The degree of accuracy, it would mean devising some form of phonetic
unfortunate thing is that so many people look upon this as an transcription, and this would make the text impossible to read
unsatisfactory state of affairs, and try to correct it. The English- without training.) In non-literary contexts, regional dialect forms.

are not common, though they are sometimes used in informal used in Britain. 'Class dialects', as they might be called, exist. They
contexts, and there are a few predictable examples, such as the are not linguistically as clearly definable as are regional dialects
differing spellings of certain words between British and American because the social correlates are not as readily delimited and de-
English. fined as regional ones - it is not simply a question of kind and
Secondly, despite the association of regional variation with degree of education. Also, English has far fewer indications of
speech, DIALECT is a term which should not be identified with position on a social scale than many other languages: in Japanese,
ACCENT. The 'regional accent' of a person refers simply to pro- for example, there are distinct, 'honorific' forms of words, which
nunciation; 'dialect', on the other hand, refers to the totality of overtly recognize class distinction.
regional linguistic characteristics - idiosyncrasies of grammar and Before going on to relate these points to the notion of style, a
vocabulary as well as pronunciation. An accent is usually the most third variable in English should be referred to, which is very similar
noticeable feature of a dialect. Whenever comedians wish to make a to those already outlined, namely, HISTORICAL variation. Our use of
joke using dialect differences, they invariably get the effect they English indicates very clearly our historical place of origin, as well
want by simply 'putting on' a new accent, and not bothering to as our regional and social background - our place on a time scale of
introduce any grammatical or other features into their speech - but some kind. Whether we like it or not, the younger generations do
in many ways an accent is the most superficial feature also. Changes not use the language in the same way as the older generations do.
in syntax and vocabulary are much more relevant for defining the This affects vocabulary for the most part, but sometimes also
differences between two dialects than are variations in pro- grammar and pronunciation. Parents' complaints about the unin-
telligibility of their children are perfectly familiar. The
Thirdly, we must remember that dialects are not just local macrocosmic counterpart to this is of course the phenomenon of
matters. My only illustrations so far have been from the dialects of language change over the centuries. 'The English language' can
one country; but far more important in a way are the dialects of hardly be restricted to that of today, but must be allowed to com-
English which operate on an international, as opposed to an in- prise earlier states of the language. Of course the boundary-line
tranational scale. Whatever differences exist between the regional between English and the language from which it came is by no
dialects of England, they have all a great deal in common when means easy to determine (it is a matter of some delicacy as to
compared with those of, say, the West lndies or the United States. whether Anglo-Saxon should or should not be included under the
The term 'dialects of English' MUST be allowed to include these heading of 'English'), but there is no doubt that SOME earlier states
areas, whose importance will undoubtedly increase as regional can be legitimately included, which is the point to be made here.
forms of literature develop. And just as there are different standards or norms for the various
But regional place of origin is by no means the only kind of regional and class dialects, so there are different norms for the
linguistic variation in a language. Just as important is the variable of historical 'dialects' also, though this is often forgotten. We cannot
SOCIAL place of origin - where we come from in terms of a position talk about Elizabethan English, let us say, in precisely the same
on a social scale of some kind. The social background of individuals terms as Modern English, or vice versa. The person who tries to
has a powerful and long-lasting effect on the kind of language they read a Shakespeare play without caring about the values that pro-
use, and there are certain general linguistic markers of class which nunciation, grammar and vocabulary had at that time is being just as
occur regardless of the particular region to which they may belong. unrealistic as the person who cries 'Preserve the tongue whiCh
For example, distinctions can often be pointed out in terms of the Shakespeare spoke!' in present-day discussions about correctness.
choices we make in the use of words referring to particular concepts Shakespearian English, as the English of any other historical
- such as how we address people or say farewell' to them, or how period, must be seen in its own terms, bearing in mind the usage of
we refer to various meals, relations, or the toilet. Terms like mate the Elizabethan period of language development, and no other.
and old man have clear social restrictions in British English. Again, Without an awareness of linguistic differences between the various
the use of 'Received Pronunciation' normally implies a degree of periods of English literature, a great deal which is of literary im-
education which need not be present for any of the other accents portance can be missed. To take just one example: without an

understanding of the normal personal pronoun system in Certain aspects of the immediate situation in which language is
Elizabethan English (the meanings of the pronouns thou and you, in used have been shown to have a strong influence on the kind of
particular), our appreciation of Hamlet's remarks to Ophelia (in Act linguistic structures which occur. One of the most important of
3 Scene 1), where there is a controlled alternation between the these is the occupational role that people may be engaged in at the
different forms of the second person, is much reduced. time of speaking or writing: the job they are doing very often carries
These three types of variation, regional, social, and historical, are with it a probability that in normal circumstances certain linguistic
very important factors in accounting for the heterogeneity of the structures will be used and others will not be. One way of speaking
English language. There are other factors too, as we shall see or writing is felt to be more appropriate to a specific professional
shortly; but these three form a group on their own. The basis for activity than another, and the members of a profession tend to
this grouping is that they are all relatively permanent, background conform in their usage to produce a consistent expression. The
aspects of any individual's use of English. Most people normally do reasons for this kind of behaviour are sometimes difficult to de-
not talk as if they were from a different area, class or time from the termine, but its extent is beyond dispute. One very clear example of
one to which they actually belong. Of course, a few people have the occupationally-motivated use of language is in the technical
ability to adopt a different dialect for humorous or literary reasons, vocabulary associated with various fields: scientists, for instance,
as we have already seen in the case of regional variation; and there make use of a range of vocabulary which precisely defines the
are also cases of people adopting what they believe to be a more phenomena they are investigating. This vocabulary does not
'educated' dialect of English in their quest for social betterment. normally occur outside of a scientific context, and alternative ways
The case of Eliza Dolittle in Pygmalion merely takes to extremes a of expressing the same ideas do not normally occur within a scien-
process which is not uncommon. But these are nonetheless the tific context - a particular substance may have a quite familiar
exceptions: on the whole we do NOT vary our regional, social or domestic name, but in the laboratory this name will tend not to be
historical linguistic norms. They are, essentially, a linguistic used, because popularity carries with it looseness of meaning, and
background against which we can make ourselves heard. They are, ultimately ambiguity. Similarly, scientists, when not 'on duty', will
to put it another way, varieties of the language on the largest not use their technical terminology to refer to everyday objects, for
possible scale. there is no need to introduce such a degree of precision into their
The relevance of these dialectal features to the study of the language. The comic situation in which a scientist asks his wife at
phenomenon of style should be clear from this paragraph: they have dinner to 'pass the H6C1206' is comic precisely because it is an
very much a NEGATIVE role to play. Regional, social, and historical abnormal, unexpected, incongruous choice of vocabulary which has
variations in a use of language have to be eliminated before we can been made.
get down to some serious study of what we consider to be 'style'. But it is not only vocabulary which characterizes an
When we talk of'Coleridge's style', let us say, we are not, in the first 'occupational' use of language - a PROVINCE, as it is sometimes
instance, thinking of his regional, etc. linguistic background; and called. The grammar is always important too. In scientific English,
people do not in fact generally make use of such phrases as 'the there are a number of constructions whose usage is different from
style of the Cockney', 'the style of Elizabethan English', and so on. other kinds of English. The way the scientist tends to make use of
Dialectal features are uncontrolled, unconscious features of our use passive voice constructions is a case in point. 'The solution was
of language; many people find it impossible to vary their usage poured .. .' is generally found in preference to 'I poured the
deliberately in these respects. Consequently, if we hope to account solution ... '. There are a variety of reasons for this, though prob-
for the relatively conscious, controlled use of language which can ably the most important is the concern to keep the account of the
produce the distinctiveness referred to above, then it must be other process being described as impersonal as possible. Similarly, legal
elements of language than these which are being manipulated. English, as found in certain documents, displays a highly distinctive
What other kinds of variation exist in English, therefore, that could and much more complex syntax that can be found elsewhere -
account for our awareness of a 'style'? unpunctuated sentences that continue for pages are by no means
A few of these other variables have been given detailed study. exceptional. And in addition to grammar, the way in which the

language is written down or spoken may be further indications of a There has been relatively little research into this field of inter-
specific brand of occupational activity. Probably the most im- personal relationships - where social psychology and linguistics
mediately distinctive feature of written advertising language is the overlap - but certain types of reasonably predictable variation have
way in which different sizes and colours of type are made use of, a been shown to exist, e.g. the different degrees of FORMALITY which
flexibility not normally seen in other written forms of the language. occur in English. It makes sense to distinguish a FORMAL from an
And a distinctive method of 'speaking an occupation', so to say, can INFORMAL style in English (with further sub-divisions within both).
be seen in the 'tone of voice' which may be adopted: those of the The kind of language we speak or write on formal occasions (such
lawyer and clergyman (while speaking in court and preaching res- as in an interview, making a speech, or applying for a job) is simply
pectively) are frequently-quoted examples, and in addition the pro- not the same as that used on informal occasions (such as in everyday
nunciations adopted by radio news-readers, political speech- conversations with our family, or writing to an old friend). This is
makers, and railway-station announcers could be cited - or indeed almost a truism. What is often ignored, however, is that the linguis-
that of most people who find themselves speaking in public as part tic features which indicate formality and informality are not just
of their professional life. There are criteria for successful and idiosyncratic, but are common to all members of the speech com-
unsuccessful uses of English in all these cases; and if we take the munity. The evidence suggests that people tend to be formal in
successful uses as a norm, then it can be shown that there are more or less the same linguistic way: they choose certain words
certain linguistic features which have a high probability of more carefully, they avoid other words like the plague, they become
occurrence on any occasion when a particular province is used. In more self-conscious over what they believe to be the 'correct'
this way, it makes sense to talk about the 'style' of a legal document, pronunciation of words, and so on. This kind of situationally-
or a political speech, as we can readily refer to the distinctive conditioned language variation, then, is yet another element con-
features in the pronunciation, spelling, grammar, and vocabulary tributing to the general distinctiveness of a use of language: a
which we would associate with these kinds of English, and which convenient way of referring to it is to call these variations of STATUS.
would not appear in the same combination elsewhere. There are other situational variables which influence the kind of
A second situational variable which conditions particular uses of English we choose to use in a given situation. For example, the
English is the relationship between the participants in any dialogue: PURPOSE for which we are using language generally produces a
this will be an important factor governing the kind of language we conventional framework or format for our speech or writing, and
choose to use. If two people are, broadly speaking, separated this can be highly distinctive. The lay-out of a letter, an
socially (as in the relationships existing between employer and advertisement, or a legal document, the organization of a lecture or
employee, student and teacher, or old and young member of a a sports commentary, are all examples of formats which have be-
family), then it is generally the case that different language come to a greater or lesser extent standardized in English. It is not a
structures will be used by the two parties, which will reflect this question of personal choice here: for a commentary or a lecture to
distinction. The socially 'inferior' person will show deference to the be successful, certain principles of 'verbal lay-out' must be
'superior' in various ways, for example by the form of address, or by followed. Then again, the broad distinction between the spoken and
avoiding the more slangy words and constructions which might be the written medium of the language has its specific linguistic
used in informally talking to social equals; and other linguistic correlates: some words and structures occur solely in speech, others
correlates can be found to indicate the dominance of the superior. only in writing. Most of the nuances of intonation have to be
Children are drilled in these conventions from an early age: 'Don't ignored in the written representation of speech, for example, and
talk like that to the vicar/Mr Jones/your grandfather ... ' is a com- most of us are well aware of the social pressures that curtail our
mon exhortation; and the emergence of social linguistic norms of freedom to write down 'four-letter words', and the like. And of
this kind can be seen in the role-playing which all normal children course the kind of language we use will undoubtedly vary de-
enter into - 'being' daddy, or the grocer, carries with it the linguistic pending on whether we speak with the intention of having our
forms of daddyness, or grocerdom, and children show remarkable words written down (as in dictation or many kinds of lecturing), or
powers of mimicry and memory in these matters. write with the aim of having our words read aloud (as in speech-

construction, news-wntmg for radio or television, drama, and, obviously, if we are dependent on someone for advancement, we
sometimes, poetry). will restrain ourselves, linguistically, and respect the conventions
It is not the purpose of this chapter to give a complete breakdown which we know are expected (e.g. the letter will be neatly laid out,
of all the categories of situationally-conditioned language which punctuation will be 'correct', formulae - such as 'yours faithfully' -
operate in English, even if this were possible in the present state of will be appropriately used); on the other hand, a letter to a close
the art. The cases so far mentioned should suffice to show the friend may carry with it all kinds of differences - loose use of
heterogeneity and fluidity of the English language. What needs to punctuation, use of slang, disregard for regular line-spacing, etc.
be emphasized, however, is that this flexibility of usage affects each Such a situation does not apply to the use of dialect features of
of us individually, and it is this which provides a crucial perspective English because, as we have seen, apart from on rare occasions, we
for understanding the question of style. In the course of one day, have little awareness of and control over their use.
each of us modulates through a wide range of varieties of English: We may summarize this discussion by saying that the English
the various levels of domesticity, professionalism, and so on, language can be seen as a complex of (to a greater or lesser extent)
through which we pass carry with them changes in the nature of the situationally-conditioned, standardized sets of linguistic variations:
language we use. The level of formality, to take but a single ex- these can be referred to as VARIETIES of the language. A variety is
ample, will vary considerably every day, ranging from the intimate therefore a formally definable, conventionalized group use of
level of family conversation (linguistically very marked, through the language which we can intuitively identifY with aspects of some
frequent use of such things as 'pet' nonsense words and slang which non-linguistic context in which it occurs (and which, as linguists, we
only the family understands) to perhaps the artificial formality of a try to formalize and explicate). An important qualification here is
chaired business meeting (with all the linguistic conventions made that we are aware of this relationship 'to a greater or lesser extent'.
use of there - proposals, secondings, etc.). What must be made Some uses of English have a very clear and direct intuitive rela-
clear - and it is this which distinguishes province, status, and the tionship to a social situation (as when the use of thou and related
like from the dialects discussed earlier - is that these distinctive forms automatically associates with a religious set of contexts);
uses of language are all relatively temporary and manipulable in other uses are much less predictable (as when an official-sounding
their use. We do not normally continue at the same level of for- phrase might have come from one of a number of different types of
mality, let us say, for a very long period of time. 'Professional' context). The concept of language variety is simply a descriptive
contexts give way to domestic interchange, which in turn may give hypothesis to account for these intuitions of formal-functional
way to a receptive appraisal of formality differences, as encountered correspondences in language; and in this sense it covers many of
on television. And, associated with this, these kinds of variation in what were above referred to as 'styles'. Phrases such as 'formal
English are all matters which we can to a very great extent control: style', 'the style of radio news-readers', and so on, are meaningful
the concept of choice is much more relevant here than it was with because it is possible to suggest clear linguistic correlates for these
the dialects. In a given situation, which has clear extra-linguistic notions.
indices of, say, formality, it is possible to exercise some degree of To say that a particular social situation has a regular association
choice as to whether appropriate, formal language is to be used, or with a particular kind of English is not to say that other kinds of
inappropriate, informal language. Of course, most normal people English may not be introduced into that situation. In principle, this
choose the former, only lapsing into the latter when they are very is always possible, for after all we can never be ABSOLUTELY certain
sure of their social ground - as, for instance, to make a joke. But the that people will behave in a maximally predictable way in a given
point is that, in principle, we have both awareness and control over a situation. But there are some language-using situations where the
number of linguistic points along the formality scale, and the possibility of making simultaneous use of a number of varieties of
question of which one to use is primarily up to us. Similarly, we all English is relatively normal, in order that a particular linguistic
know the conventions for letter-writing; but we may choose to effect be achieved. Literature and humour are the clearest examples
ignore them if we so wish. Whether we do so will depend almost of this happening, but cases of 'stylistic juxtaposition' can be found
entirely on our relationship to the person we are addressing: elsewhere too. For example, a political public speaker may intro-

duce quotations from the Bible into the oration to point a particular non-idiosyncrasy: we cannot recognize the individuality of authors
issue; or a television advertisement may introduce language from a until we are first aware of the language habits of their time, i.e. the
scientific form of English in order to get some of the scientific linguistic features of the various dialects, provinces, and so on,
overtones rubbed off onto the product; or a sermon may introduce against which background they can display themselves. And this
television advertising jingles to make an idea strike home more means that any study of individuals requires the prior recognition of
directly. These are reasonably frequently-occurring examples of the more general linguistic usages contemporaneous with them.
language from two or more varieties being used in a single situation, (This explains the difficulty of trying to identifY the authorship of
and the kinds of juxtaposition which occur are to a certain extent texts in languages which are no longer spoken - as in the case of the
predictable, especially when compared with the essentially unpre- Pauline epistles. Whether the linguistic idiosyncrasies of the
dictable juxtapositions which are introduced into literature and epistles are those of one man [= Paul?] or not depends on whether
humour. Many kinds of joke are successful because they introduce we can first eliminate from the discussion those features common to
incongruity of a stylistic kind into the punch line; and in literature, it other letter-writers of the period, and those common to the
is a standard procedure for an author to incorporate into a work language as a whole at that time. And in view of the fact that there is
snatches or even extended extracts from the non-literary varieties of so little comparative material extant, it is doubtful whether the
English. It is difficult to see how this could be otherwise, but some problem is solvable.) In one sense, then, linguistic idiosyncrasy is
authors go in for stylistic borrowing of this kind much more widely subordinate to the study of shared uses of English; and this is of
than others: the chiaroscuro of overtones and association in much course the position taken by those who are engaged in teaching the
of James Joyce is to a very great extent explicable in terms of other language, where they are in the first instance trying to teach the
varieties of (particularly religious) English; and T. S. Eliot is language 'as a whole', and disregarding those features which belong
another who constantly makes use of this technique in a very to individuals. But from the point of view of the study of style,
definite way. Moreover, many of the so-called 'revolutions' in the idiosyncrasy - as some of the viewpoints outlined at the beginning
use of poetic language in the history of English literature can of this chapter suggest - becomes of primary importance.
ultimately be reduced to attempts to replace the methods of expres- One thing must be made clear at this point. By IDIOSYNCRASY I
sion associated with one variety of the language by those associated am not referring to those uncontrolled, and normally uncontrollable
with another: an example would be the introduction of scientific features of our spoken or written utterance which are due entirely to
language by the Metaphysical poets, or by some twentieth-century our physical state and which will always be present in everything we
authors, into a poetic context where scientific language had been speak or write. In everything we say, there will always be an idiosyn-
almost completely absent for some time. Whether the 'language of cratic voice-quality, a background vocal effect which identifies us as
the age' is or is not the language of poetry is not a matter for individuals, and this we do not normally change (unless we are
discussion here; but it should be noted in passing that this argument professional actors or mimics, of course). The analogue to voice
will never be resolved until an attempt is made to clarifY the notion quality in the written medium is our personal handwriting.
of 'language of the age' as such - and in order to do this, SOME Similarly, if we speak with a particular kind of speech defect, or
reference to a theory of language variety is going to be necessary. using some psychopathologically-induced set of recurrent images,
So far I have been discussing aspects of language variation which these may well be idiosyncratic, but this too is a different sense from
are basically group uses of language. The remaining factor that intended by the concept of stylistic idiosyncrasy. In the latter
accounting for linguistic heterogeneity stands apart from all these, case, I am referring to the linguistic distinctiveness individuals can
in that it is concerned with the language habits idiosyncratic to a introduce into their language which is not shared by other members
person, those which distinguish someone from the other members of society (i.e. not a variety) and which is capable of conscious
of a group, as opposed to integrating that person linguistically with control. The author of the language may choose to put something in
them. In one sense, of course, linguistic idiosyncrasy is less im- or leave something out. The important word here is 'may', as very
portant than the dimensions of variety outlined above, as we can often, depending on someone's experience of using the language,
only be aware of idiosyncrasy against a background of specific linguistic indices of personality may make their appearance

with apparently no conscious effort on the author's part. We are all technique of comprehensive analysis, so that, once it has been
familiar with the linguistic idiosyncrasies of certain public figures or mastered, students of language may find it easier to appreciate the
of favourite authors; we talk about an author's name being 'stamped complexity of language use.
indelibly on every page', and so forth. But in principle this is This now brings me to the final aspect of stylistics which I want to
something over which authors have a large measure of control: they discuss here, namely, What ARE the techniques whereby the 'style'
can change words, alter their order, add and delete at will. Theirs is of a text can be analysed? The kinds of language variation which
the decision which ultimately controls what we see or hear, and may be found in any piece of language, we must remember, reduce
which ultimately defines their individuality in the use of language. to three basic types: there are the features I have called DIALECTAL
The linguist's job here is to identifY and explain the idiosyncratic (regional, class, historical) which partition the English language in
effects which authors have introduced into their use oflanguage, to terms of one set of dimensions; cutting across these, there is a
see whether these form any kind of pattern, and to try to demon- second set of dimensions, relating to specific factors in SOCIAL
strate their purpose in relation to the work as a whole. SITUATIONS, such as occupation, relative status, and purpose; and
It is important to emphasize, once again, that linguists do not thirdly, there is the possibility of IDIOSYNCRATIC variation, which
have an evaluative role in this matter: theirs is, basically, a des- allows for the modification of the group norms by individual users.
criptive task. They are not studying an author's work to decide It needs a fairly sophisticated stylistic theory to be able to account
whether it is good or bad, representative of this quality or literary for every factor; but from the point of view of specifYing a procedure
tradition or that: its 'place' in literature is not of primary importance for analysis, ALL these dimensions of language variation can be
to them AS LINGUISTS - though of course this may well have entered studied in precisely the same way, using any of a number of possible
into their decision as to which text to analyse in the first place, a techniques suggested by General Linguistics. Exactly which tech-
decision not made on linguistic grounds. Linguists are primarily nique we use will of course be the outcome of our particular
concerned with ensuring that all features relevant to the training and predilections, and of the specific theory of language
identification of an author's own behaviour are understood. If some structure we may adhere to. These days, there is a great deal of
features are omitted through ignorance, they would argue, then controversy as to which of the many linguistic theories available
there is a very real danger of relevant information for the overall provides the best basis for the analysis of any given piece of
qualitative assessment of the author by the critic being overlooked. language, but the existence of certain features, or LEVELS of
The reason why STYLISTICS, the linguistic study of what is con- language structure, seems to be generally recognized; consequently
sidered to be 'style', has become so popular over recent years, it it is probably easiest to illustrate the kind of preconception a linguist
would seem, is precisely that, using the traditional methods of might bring to bear in studying a text from the stylistic point of view
language analysis and literary criticism, so much of importance for by outlining what is involved in these levels. The most useful levels
this basic assessment DOES get overlooked. Students ofliterature, or of structure to recognize for stylistic purposes have already been
of any use of English, frequently begin their analysis of a text in a discussed in the earlier part of this book: phonetics, phonology,
highly impressionistic way, relying on their innate sensitivity to grammar, and vocabulary. (The concepts of phonetics and
produce the results they seek. But sensitive response alone is - phonology are primarily reserved for the study of speech: for the
apart from very rare cases - an inadequate basis for reaching a clear study of a written text, the analogous levels could be referred to as
understanding of the message which is being communicated. Most GRAPHETICS and GRAPHOLOGY respectively.) I would argue that the
people do not have the ability to approach the study of the language distinctiveness of ANY text can be broken down in terms of these
of a text in any systematic, objective kind of way. The gifted few, it levels: whatever distinctive stylistic feature we may encounter in
is true, may be able to sum up the relevance of a poem for them English, it can be described as operating at one or some com-
without entering into any systematic procedure of analysis, but for bination of these levels.
the majority, the initial aesthetic response needs to be To obtain a clearer picture of what is involved, I shall illustrate
supplemented by some technique which will help to clarifY the the kind of distinctiveness which might occur at each level, taking my
meaning of a text. Stylistics, then, hopes to provide just such a examples primarily from literary texts. At the PHONETIC level would

be studied any general features of sound which help to characterize a certain extent overlap with non-linguistic considerations (e.g.
a text, such as when a particular voice quality (or set of qualities) is matters of colouring), but the point is that from the stylistic point of
associated with a particular use of language (as in much religious view, even such non-linguistic matters as choice of colour might
and legal professional speech). The 'clerical' voice is a well-recog- have a contribution to make to the definition of the distinctiveness
nized phenomenon, and this principally refers to a quite different of a particular use of language - and thus would have to be allowed
'set' of the vocal organs from that normally used by the clergyman in for in any stylistic theory.
everyday conversation. Also under phonetics, one would consider The PHONOLOGICAL and GRAPHOLOGICAL levels are easier to
those aspects of speech which would normally be referred to under illustrate as they relate to more familiar matters. There are, broadly
the heading of SOUND SYMBOLISM - a hypothesized capacity of speaking, two areas of potential distinctiveness: what I would refer
sounds to intrinsically reflect objects, events, and so on, in real life. to for speech as the SEGMENTAL and the NON-SEGMENTAL areas.
This view may be illustrated by people who claim that there is Segmental characteristics of style would cover the use of specific
something in the nature of an [i:] sound, for instance, which makes vowels and consonants within a particular language's sound system
it necessarily relate to smallness in size, or whiteness, or something in combination in a distinctive way, as when we make use of
else; or that onomatopoeic words - such as splash or cuckoo - could reduplicative effects such as alliteration, assonance, and rhyme in
have no other shape because they contain the sounds of real life English. It is important to note that these devices have a major
('biscuits are so called because of the sound they make when you structural, as well as aesthetic, function - that is, they are the
break them'). These arguments have been generally shown to be province of phonology, as opposed to phonetics. Alliteration, for
unfounded. Even such clearly onomatopoeic words as splash vary in example, may well have an important aesthetic appeal; but from the
their form from one language to the next, showing evidence of point of view of its overall function in a poem, it has an equally
non-naturalistic influence; and there are always counter-examples important - and sometimes a more important - role as an organ-
to any generalization we might care to make about the 'inherent izing process, linking words more closely than would otherwise be
meaning' of sounds such as [i:]. But it is nonetheless the case that the case. For example, when we read such a line as 'Thron'd in the
various uses of language (poetry being the clearest example) do try centre of his thin designs' (from Pope's Epistle to Dr Arbuthnot), the
to make use of speech sounds in as evocative a way as possible. If major function of the alliteration is to force the words 'thron'd' and
poets consider a particular sound to have a powerful atmosphere- 'thin' together, and thus produce a juxtaposition of the concepts
creating potential, then they may well make use of it (i.e. words 'mediated' by the words, which in the present context produces an
containing it) more frequently than usual. Of course we have to ironic contrast. This kind of thing is presumably one of the factors
remember that in general we can only interpret sounds in a given underlying phrases such as 'fusion and meaning', or when we talk
way once we know the theme being expressed by the words: [s] about a poet's 'intensifYing' meaning. And similar illustrations
sounds in a poem about a swan may well reflect the noise of the could be found for the other re duplicative segmental processes. In
water, but in a poem about evil might equally appropriately be passing, we should note that it is difficult to generalize about
intended to conjure up the noise of serpents, and the like - in other phonological distinctiveness for more than one language. Such
words, there is no 'general meaning' for the [s] sound in language, matters as alliteration and rhyme are essentially deviations from the
or even in English. But having said this, we may still plot the way the normal ways of distributing consonant and other phonemes in
poet manipulates specific sounds, seen as individual, atmosphere- English. That is why these effects are so noticeable: they are not
setting sonic effects, to reinforce a particular theme, and this would normally encountered in our contact with English. In a language
be studied at the phonetic level of analysis. In the written medium, where initial reduplication of phonemes WAS normal, however -
we would be referring on similar grounds to sueh matters as the where prefixes were the routine way of indicating cases, for instance
general size and shape of the type being used (as in the dis- - then much less effect would be gained by alliteration, and we
tinctiveness of posters, newspapers), and the lay-out of a text on a could anticipate that other phonological features than this would be
page (as when Herbert writes a poem about an altar in the shape of used to produce dramatic and other effects. Similarly, in a language
an altar). The PHONETIC and GRAPHETIC levels of analysis, then, to like Latin, where - because of the inflectional endings - it is

difficult NOT to rhyme to some extent, we do not find rhyme being In English, variations in spelling for special effects are uncommon,
used as a literary device with anything like the same frequency as in though we do find archaic spellings introduced into poetry, or
English. notices printed in an old-fashioned way. An example of this, again
The other aspect of phonology is the non-segmental; that is, the from Eliot (East Coker), is a good example of the impact of the visual
features of intonation, rhythm, speed, loudness of articulation, and medium which could not possibly be translated into spoken form:
other vocal effects we introduce into speech in order to communi-
cate attitudes, emphasis, and so on. Spoken English is highlydis- And see them dancing around the bonfire
The association of man and woman
tinctive from this point of view. Taking intonation patterns alone,
In daunsinge, signifYingmatrimonie -
there would be good grounds for distinguishing between most
A dignified and commodious sacrament.
varieties of spoken English currently in use. There is the charac- Two and two, necessarye coniunction,
teristically wide range of pitch movement in the public-speaker as Holding eche other by the hand or the arm
opposed to the narrower range in everyday conversation; the Whiche betokeneth concorde.
'chanting' effect of the sermon; the restrained, regular movement of
the news-reader; and so on. And when we consider features other More commonly in English, we find variations in punctuation, even
than pitch, our classification can become very precise: compare the to the extent of occasionally omitting this altogether (as in the final
varying speed and loudness of the sports commentator with the pages of Joyce's U/ysses, for example).
measured speed, loudness and pause of the professional reader; the Vocabulary, a language user's 'choice of words', or DICTION, as it
many vocal effects (such as increasing and decreasing the tension of is sometiII).es called, is presumably so familiar an aspect of a per-
the muscles of the vocal organs for stretches of utterance, which son's style that it does not need detailed illustration here. At this
produces a tense, 'metallic' effect and a lax effect respectively) level, stylistics tries to determine the extent to which certain words,
which are introduced into the use of English for television combinations of words, and types of word are part of the dis-
advertising; or the primary role of rhythmic variations in estab- tinctiveness of a use of language. All varieties of English make use
lishing the linguistic basis of poetry. It should be clear from these of a restricted kind of vocabulary, e.g. the learned, technical
examples that a great deal of our awareness of stylistic dis- vocabulary of scientific English, the loosely colloquial vocabulary of
tinctiveness in speech derives from the perception of 'prosodic informal conversation, the formal, precise vocabulary of legal
features' of this kind. When we vaguely hear speech in the distance documents, the archaic vocabulary of much of religious English,
and say 'That sounds like .. .', we are generally basing our and so on. Sometimes a variety can be identified merely on the basis
judgement on the dominant prosodic variations we can hear. of certain items of vocabulary, as in the use of such words as
The analogous features to phonology in the writing-system of a heretofore, which is used only in legal English or attempts to simulate
language can be roughly summarized as the spelling and punctuation it. More often than this, however, a style is lexically distinctive due
of that language. I say 'roughly', because a great deal more is covered to certain words being used more, or less, frequently than in other
by graphology than is traditionally understood by these labels, e.g. the varieties or individuals - authors may be said to have their
difference between upper and lower case symbols is of systematic 'favourite' words, for instance. Or there may be a particular dis-
importance in English (and not just a matter of aesthetic appeal): it can tribution and proportion of various categories of word in a text, e.g.
be used distinctively, as when we write something out in capitals to the highly distinctive mixture of technical, slang, formal and infor-
achieve extra prominence, or when we introduce a graphological mal vocabulary in sports commentary, or the parallel use of tech-
change in order to indicate a change in context (without actually nical terms and non-technical glosses in many kinds of lecturing.
having to say so), as when Eliot writes Again, an individual may produce stylistic effects by coining new
words (e.g. theirhisnothis,Joyce) or by putting unexpected words in a
I didn't mince my words, I said to her myself, standardized context, as in Thomas's a grief ago (where the ex-
Now AIbert's coming back ... (The Waste Land) pectation of a noun of time imbues the notion of grief with temporal
associations) and similar examples. The choice of a specific word is

one aspect of style; the placing of that word in a specific context is structions has a very restricted range of sentence structures at its
another, quite different aspect. disposal (high frequency of imperatives and imperative-like
In studying vocabulary, we are of course studying meaning to a elements, absence of questions). Newspaper reporting generally
certain extent, but meaning is not restricted to single words or small makes use of relatively short, uncomplicated sentences. News-
combinations of words. Of relevance for stylistic study is the way in reading (and most other forms of radio narrative) never uses any-
which the overall meaning of a use of language is organized, and it thing other than statements. In scientific English, equations and
is this more general study of meaning' which takes place at what is formulae can replace elements of sentence structure and sometimes
often referred to as the level of DISCOURSE. For example, when we whole sentences. In commentary, conversation and advertising,
talk about the 'theme' of a poem or novel, or discuss the 'pro- there is a very frequent and varied use of MINOR sentences (i.e.
gression' of ideas in a play or an income tax form, we are referring structures which function as sentences, but which do not have the
to the most general patterns of meaning that we have been able to subject-predicate structure characteristic of the majority of English
discern in a text, and there is a great deal of stylistic significance to sentences - as in hello, sorry, and so on). The traditional distinction
be said here, if this is done systematically. The discourse organ- between SIMPLE, COMPOUND, COMPLEX and MIXED sentence types
ization of a lecture, for example, with its steady development inter- is relevant for categorizing the kinds of distinctiveness we find in
spersed by passages of recapitulation and anticipatory summary texts, and these categories can be further subdivided - the number
(e.g. 'there are three things I'd like to say about this .. .') is quite and type of subordinate clauses, for example, varies considerably
different from the regular, alternating flow of descriptive narrative from variety to variety as would be clearly shown by a comparison
and background comment which characterizes a sports com- of political public speaking (where they are very frequent, tending to
mentary, and this is different again from the near-random pro- pile up on each other in rhetorical climaxes) with radio news
gression of ideas in conversation. I take my examples here from the broadcasting (in the latter, subordinate clauses are common, but
less familiar (spoken) varieties: in the written medium, the concept their distribution is more sporadic, and they rarely are used in
of the paragraph, which is a semantic unit (cf. the notion of 'topic anything approaching a 'cumulative' way). In literature, changes in
sentence', and so on), has long been with us, as have such visualist the direction of the plot, or the theme, can be indicated by altering
devices as sub-headings, spacing variations, and diagrams, which the kind of sentences generally being used; and this device is of
make the movement of thought relatively unambiguous and easy to course extremely common as one index (often the most noticeable)
perceIve. of character - Dickens, for example, regularly gives his characters a
Finally, there is the grammatical level of analysis, which is prob- predictable linguistic basis, and sentence structure usually has an
ably the most important component of any stylistic description. important distinctive role to play in this. Again, the absence of clear
There is invariably more to be said about the grammar of a text than sentence boundaries may be a major way of communicating a
about any other level, and in order to make a successful study here particular effect, as with certain stream-of-consciousness tech-
it is essential to have fairly clear ideas about the general nature of niques. And there is a great deal else which can be manipulated to
English grammar, as suggested by some grammatical theory. It is make sentences work in a. distinctive way (e.g. the devices that may
impossible even to outline what would be involved in a complete be introduced in order to LINK sentences to each other, such as
grammatical description of a text here: some further reading on this cross-referencing, repetitions of words, the use of adverbs like
question is given in the Bibliography on p 349. But if we consider however and conjunctions).
merely the kind of variations which occur at ONE point in English This has been a very brief outline of a possible method of
grammar, it might be possible to get an impression of the overall discovering some principle(s) of organization in the mass of
complexity involved, and so not underestimate the scope of linguisitc features which constitute the distinctiveness of a use of
grammatical analysis. The TYPE OF SENTENCE one may find in a text English. It sh0uld be clear that ALL levels of analysis enter into this
is often a reasonably unambiguous diagnostic indication of its pro- distinctiveness, though some (the grammatical and lexical in par-
venance. There is nothing like the long complex sentences of legal ticular) have a more dominant role on most occasions. The concept
documentation elsewhere in English. The language of instructions of 'style' which emerges from this approach, when seen within the

perspective of language varieties as presented earlier, is thus very

much a cumulative, developing, dynamic one: it is essentially a
descriptive convenience which summarizes our awareness at any
given moment of the controllable linguistic features that distinguish
one use of English from any other. The specification of these uses is
in terms of the dimensions of variation outlined in the first half of
this chapter: the features are identified and inter-related in terms of
the levels of analysis outlined in the second half. It is in such
attempts to provide a relatively objective way of talking about and
analysing language variation systematically, precisely, and compre-
hensively, that linguistics hopes to be able to make a permanent
contribution to the study of English style.